ROOFTOP GREENHOUSES: LCA AND ENERGY SIMULATION

K. Benis1; R. Gomes1; R. Vicente1; P. Ferrão1 ; J.E. Fernández2

1: IN+ Instituto Superior Técnico, Av. Rovisco Pais 1, 1049-001 Lisbon
2: MIT Department of Architecture, 77 Massachusetts Avenue, Cambridge, MA 02139-4307

ABSTRACT

Building-Integrated Agriculture (BIA) has the potential to offer a new dimension to our
buildings, providing locally grown food that increase urban resilience. There are two main
forms of BIA: Rooftop Greenhouses (RG) and Vertically Integrated Greenhouses (VIG). This
paper focuses on RG, i.e. setting up hydroponic greenhouses on top of flat roofs. With 85% of
Lisbon’s building stock built before 1980, when there were no insulation requirements, there
is a strong retrofitting potential using RGs. This should be considered together with the
energy requirements of hydroponic environments, particularly for indoor temperature control.
This work combines Life Cycle Assessment (LCA) and energy simulation of a RG
implemented on a residential building located in Lisbon. The analysis is aimed at quantifying
the environmental impact, but also the energy requirements of the RG through its operation
phase. The effect of the RG on the indoor temperature of the last floor apartments was
analyzed. The first results show an improvement of the indoor temperature in the winter
period and an undesirable increase in the indoor temperature during summer. These results
highlight the need to evaluate different scenarios such as recovering part of the cooling loads
used in the greenhouse and transferring them to the building, the application of insulation in
the rooftop slab and the evaluation of night ventilation.
The aim of this study is to constitute a first step towards a quantitative basis for decision-
making in the implementation of RGs in building retrofit interventions, by showing what
alternatives would be most effective in delivering CO2 emissions reductions, along with their
respective costs and amounts of saved energy —thus offering an indication of which option is
to be favored to guarantee sustainability and cost-effectiveness.

Keywords: Building-Integrated Agriculture, Rooftop Greenhouse, LCA, energy simulation,

indoor temperature, controlled environment

INTRODUCTION
Building Integrated Agriculture (BIA) consists of the application of hydroponic greenhouse
methods adapted for use on top of or in buildings [1]. This study analyzes the implementation
of a RG for lettuce and leafy greens production on a low-rise multi-family dwelling located in
Lisbon, with 18 apartments and 60 estimated inhabitants.
Lettuce production systems use Nutrient Film Technique (NFT), a system where re-circulated
nutrient solution is pumped from a reservoir to slopping polyethylene or PVC channels, in
which plant roots are placed in planting holes separated by a distance of 15 to 30 cm.
The RG occupies the whole area of the flat rooftop (i.e., 270m2 with a production area of
225m2) with 26 plant sites per m2, which provide a yearly yield of 16,85 tons (threefold of the
demand of the building’s inhabitants). Sizing characteristics were adapted from information
provided by local growers. The 21 identical existing buildings in the neighborhood offer the
possibility of diversifying hydroponic cultures to cover local needs.

CISBAT 2015 - September 9-11, 2015 - Lausanne, Switzerland 95

 LIFE CYCLE ASSESSMENT

Goal and Scope

The goal of this LCA is to quantify the environmental impacts of rooftop greenhouse
hydroponics production systems in a residential building in the city of Lisbon. The functional
unit is 1 ton of greenhouse food produced (lettuce and leafy greens). The system boundary was
defined as in Figure 1 (cradle-to-gate). The LCA was modeled on SimaPro, using the EcoInvent
3.1 database. The model includes five main processes: (1) greenhouse structure; (2) electricity;
(3) water use; (4) growing process; (5) waste management.

Figure 1: Life Cycle Inventory of greenhouse lettuce production system (hydroponics)

Life Cycle Inventory

The greenhouse structure is made from steel, aluminum and polycarbonate. Hydroponics
growing channels are made of polyethylene. For this study, a leading Portuguese greenhouse
manufacturer provided information concerning the characteristics and quantities of materials
used in a “standard” greenhouse, which were adapted to this particular case. Local growers
provided the specific dimensions and quantities of hydroponic equipment for growing lettuce.
The electricity process includes total electricity consumption (kWh) for the entire greenhouse
activities (i.e., pumping systems, ventilation and lighting).

96 CISBAT 2015 - September 9-11, 2015 - Lausanne, Switzerland

 The water use process includes total water consumption (m3). Whereas conventional
agricultural production requires 120 liters of water per kg of lettuce, water use efficiencies in
hydroponics are usually around 20 liters per kg of lettuce [2]. Based on data provided by local
growers, the calculations for this case study led to a result of 19,23 liters of water per kg of
lettuce.
In the growing process, the production of seeds was not considered because of lack of data. The
growing medium process includes material and energy inputs for the manufacturing of substrate
(i.e. rock wool), and its packaging (plastic and cardboard). The fertilizers process includes
building infrastructure and electricity needed for the production of fertilizers. The pesticides
process considers the production of pesticides including materials, energy use, and
infrastructure.
Electricity consumption rates, water use, and quantities of seeds, growing medium, fertilizers
and pesticides used per kg of production were also obtained from local growers. The Portuguese
electricity mix was used in the model.
Different waste scenarios were modeled depending on the nature of the waste: (1) greenhouse
structure, namely construction materials and hydroponics equipments (steel, aluminum and
plastic); (2) organic (plant roots and waste); (3) inorganic (rock wool); (4) plastic and
cardboard packaging. The RG construction materials, excluding plastic, were assumed to have
a lifespan of 25 years. For the roof and walls composed of polycarbonate, the lifespan
considered was 10 years. For hydroponic polyethylene equipment (i.e. channels and pipes), the
lifespan considered was 4 years. Distances of transportation were considered, as well as GHG
emissions from the waste treatment process. In LCA studies of greenhouse food production, the
cut-off method [3] is the most commonly applied for the allocation of compost and recycling
process: only loads directly caused by a product are allocated to it. Thus, composting of organic
waste and recycling processes (for metals and plastics) were excluded.
All processes required transportation from production sites to the greenhouse, and were
calculated using the formula: t x km. Distances traveled were based on the discussions with
local hydroponic lettuce producers, regarding the locations of their suppliers.

Figure 2: Life Cycle Inventory of greenhouse lettuce production system (hydroponics)

Life Cycle Impact Assessment

Environmental impacts of hydroponic production of lettuce and leafy greens in the RG are
shown in Figure 2. The production process (i.e. electricity, water use and growing process) has

CISBAT 2015 - September 9-11, 2015 - Lausanne, Switzerland 97

 a significant share in most of the categories (ReCiPe Endpoint (H) Assessment Method,
Europe). Among the growing process components, pesticides are the major contributors to the
impacts. Greenhouse structure and waste have less environmental impacts, since most of the
materials are to be recycled at their end of life.

BUILDING AND GREENHOUSE SIMULATION

The building energy modelling allows performing an initial assessment of the thermal needs of
the building and of the greenhouse during the operating phase and the possibility to evaluate
different solutions for its acclimatization.

Simulation inputs of the building

The energy simulation of the building was performed using the software Energy Plus version 8
and the geometry was defined using Google Sketchup. It is important to note that this simulation
needs to be calibrated with in situ measurements to better represent the energy performance of
the existing building. Therefore, the results presented here are a first analysis of the building
thermal needs.
For this simulation, the building zoning was done considering spaces with different uses (i.e.,
kitchen, rooms and living rooms). The building was constructed in 1960 and the constructive
solutions defined in the simulation were a double brick wall with air space for the exterior walls
and a precast concrete joist and brick panel for the slabs.
The windows defined in the simulation are constituted by a clear 6mm glass installed in an
aluminum frame, with external plastic shutters. Internal gains of the building were defined,
namely occupation, lighting and equipment, considering predicted and reasonable values for the
building typology. The air infiltration values were defined accordingly to specific bibliography
[4] but are expected to have more accurate results in the future with in situ measurements.

Simulation inputs for the greenhouse

One main purpose for considering a greenhouse structure in a building rooftop is to create a
controlled environment in terms of temperature and humidity for optimum growing conditions
within a predictable and repeatable time schedule when compared to growing outside in a non-
controlled environment. Considering the greenhouse structure, construction materials and
design, it can become too warm in the summer and cold in the winter which could affect the
crop production. The best indoor conditions control systems should not only be effective in
providing the desired environment, but also be designed to be unobtrusive within the
greenhouse system. Evaporative cooling is a common way to reduce indoor temperatures for
greenhouses in dry climates [9] and basically consists of a process that reduces air temperature
by water evaporation into the airstream. As water evaporates, it absorbs energy from the
surrounding environment (greenhouse) decreasing the temperature of the air flow. Fan and pad
evaporative systems consist of exhaust fans at one end of the greenhouse and a pump circulating
water through and over a porous pad installed at the opposite end [5, 6, 7, 8]. The cooling
efficiency is dependent of the pad wall material (corrugated cellulose, aspen pads or aluminum
and plastic fibers) and air flow velocity and can vary between 70 to 80% [6, 9]. Additionally,
the outside air conditions, namely the relative humidity and temperature, affect the cooling
potential of the pad wall system [8, 9].
Evaporative pad cooling system
The ventilation sizing for the evaporative pad system considered in this greenhouse was
performed considering the air flow value of 2.4 m3 min-1 per m2 of floor area [8]. Considering
the greenhouse geometry it was considered that the system has three fans, one for each zone

98 CISBAT 2015 - September 9-11, 2015 - Lausanne, Switzerland

 considered in the simulation of the greenhouse. The pad wall considered is constituted by
corrugated cellulose since this is the most widely type used for evaporative pad walls [8]. The
pad wall was considered to be in the north façade of each zone since this is the direction of the
prevailing winds in Lisbon [11], increasing the efficiency of the pad system. For heating
purposes it was considered an electric baseboard equipment to heat the greenhouse. The indoor
temperature setpoint defined for the greenhouse was 24-28ºC.
Results
As it was expected considering the building typology and the constructive solutions, there are
significant heating and cooling needs in all the apartments, as the number of annual hours with
indoor temperatures above 26ºC in the summer period and below 18ºC in winter period is
considerably high (considering no HVAC systems). The effect of the rooftop greenhouse in the
building indoor temperature can be observed in the figure 3. As the temperature of the
greenhouse was defined to be between 24 and 28ºC during all year it can be observed that the
temperature of one room in the last floor increased with the implementation of the RG. This is
a result from an increase of the heat gains from the greenhouse considering the low thermal
resistance of the existing rooftop slab. Although this can be considered positive in the winter
period, it represents a thermal comfort disadvantage in the summer period. One possibility to
overcome this situation could be to improve the slab insulation or increase the night ventilation
on the building.

Figure 3: Building simulation model (southeast view) and annual indoor temperature in one
room of the last floor before and after the rooftop greenhouse

Simulation limitations
Several limitations regarding the simulation of the building and of the greenhouse can be
highlighted. The calibration of the model with in situ measurements in the existing building,
energy audits, occupation patterns evaluation and other relevant parameters, will contribute to
a more accurate building simulation and a better analysis of the operative phase of the building
with and without the greenhouse. Regarding the greenhouse, it is relevant to highlight the
possibility of the existence of a gradient of air temperature between the pad wall and the fans
(not considered in the simulation). In fact it is expected that the temperature near the fans will
be higher than on the opposite side of the greenhouse. Other relevant aspect to be analyzed is
the effect of the wind on the pad evaporative system. The wind profile specific from the
building location will contribute to this analysis.

ONGOING AND FUTURE WORK

In a next step, the energy modeling of the greenhouse will allow for the assessment of its
heating and/or cooling energy consumptions, which will constitute an additional process of

CISBAT 2015 - September 9-11, 2015 - Lausanne, Switzerland 99

 this LCA model, increasing the environmental impact of the use phase of the RG. A sensitivity
analysis looking at reducing the impacts of the major contributors to the environmental
impacts will be performed. This assessment will rely on the analysis of different scenarios
such as the evaluation of different greenhouse acclimatization solutions, building insulation
application, and passive solutions in order to reduce LCA impact of the greenhouse together
with existing building. For all the scenarios, the energy savings will be calculated but will also
be considered the input material such as ducts, fans, heat recover units and other materials and
construction works. Also, the possibility of implementing a photovoltaic system to provide
energy to the greenhouse will be considered and analyzed from a LCA perspective.
Besides, the size of the greenhouse and consequently the crop production should be evaluated
in order to define the most suitable solution regarding global environmental impact. The main
goal is to achieve the best scenario that includes solutions for the building as well as for the
greenhouse.
A Life Cycle Cost (LCC) analysis performed in parallel to the energy flows scenarios will lead
to the constitution of a quantitative basis for decision-making, by showing what alternative
would be most effective in delivering CO2 emissions reductions, along with their respective
costs and amounts of saved energy —thus offering an indication of which option is to be
favored to guarantee sustainability and cost-effectiveness.
REFERENCES
1. Puri, V., Caplow, T.: How to grow food in the 100% renewable city: Building-integrated
agriculture. In P. Droege (ed.). 100% Renewable: Energy Autonomy in Action. Earthscan,
London, UK, pp 229-241, 2009
2. Kratky, B.A.: Three non-circulating hydroponic methods for growing lettuce. Proceedings
of the International Symposium on Soilless Culture and Hydroponics. Acta. Hort. 843:65-
72, 2009
3. Ekvall, T., Tillman, A.M.: Open-Loop Recycling: Criteria for Allocation Procedures.
International Journal of Life Cycle Assessment, 2 (3), pp 155-162, 1997
4. Villi, G., Peretti, C., Graci, S.: Building leakage analysis and infiltration modeling for an
Italian multi-family building. Journal of Building Performance Simulation, 2012
5. Caplow, T., Nelkin, J.: Building-integrated greenhouse systems for low energy cooling.
New York Sun Works, USA, 2007
6. Strobel, B.R., Stowell, R.R., Short, T.H.: Evaporative Cooling Pads: Use in Lowering
Indoor Air Temperature, Food, Agricultural and Biological Engineering,
http://ohioline.osu.edu/aex-fact/0127.html, consulted in May 2015
7. Bucklin, R.A., Leary, J.D., McConnell, D.B., Wilkerson, E. G.: Fan and Pad Greenhouse
Evaporative Cooling Systems, University of Florida, Florida, 2010
8. The Center for Agriculture, Food and Environment. Fan and Pad Evaporative Cooling
Systems. https://ag.umass.edu/fact-sheets/fan-pad-evaporative-cooling-systems, consulted
in May 2015
9. Giacomelli, G.A.: Evaporative cooling system: pad and fan, Controlled Environment
Agricultural Center, The University of Arizona, USA, 2003
10. Franco, A., Valera, D.L., Peña, A.: Energy Efficiency in Greenhouse Evaporative Cooling
Techniques: Cooling Boxes versus Cellulose Pads, Energies 2014
11. Windfinder, http://pt.windfinder.com/windstatistics/lisboa, consulted in May 2015

100 CISBAT 2015 - September 9-11, 2015 - Lausanne, Switzerland